jenders

This

Spend any amount of time talking to customers and you may find that many of them, maybe even most, don't want the responsibility of maintaining infrastructure. They want "insert money, get more software." The few that run their own infrastructure often come back to SaaS after realizing they made a monkey's paw wish. Many aren't ready for, or interested in, that level of involvement and responsibility. It's a burden, and the economies of scale available to software make marginal scaling more efficient when done by the provider. Not always, but it's often a perfect fit.

But only a bad salesman dictates what the customer needs— talk to your customers.

wrt (5): Is it not good for the user base at large or is it not good for users who choose to live as nomads on a boat?

I feel like we have to be realistic about the problems we’re solving here—-is the goal to optimize for 95% of users or 100%? If 100%, I have to ask if the tradeoffs for the last 5% of users are worth the squeeze to the 95%? SaaS really is the right solution for many people; not all but many.

I’m not advocating for anything but for distros to stop recommending swap partitions. I advocate for right-sizing your hardware to predicable workloads and not using swap at all. On your pet servers or desktops where the workload is experimental or wildly unpredictable, swap files make more sense over partitions.

This is a mechanism post not a policy post. If you want to see my take on swap, see my last response on that gift.

Swap is not recommended on servers where responsiveness and determinism are more important than failure recovery of any one server.

I love this kind of stuff. Great share

Looks like this was added in util-linux 2.40. Works with btrfs correctly too! Added.

This recommendation is based off the current implementation.

See:

https://github.com/torvalds/linux/blob/827751b699b79a6e56998...

https://github.com/torvalds/linux/blob/827751b699b79a6e56998...

The thread is quoted to show that they have been functionally equivalent for several decades and yet distros continue to push outdated recommendations.

+1. This test and its conclusions are flawed. See CAV/CLV

you love to see it

What subsystem are you referring to?

I wrote this years ago before LLMs made it easy to validate guesswork. I’ll update! Good call

Ah the old HN hug. I’ll fix it

Fantastic

>> ZFS is overrated and mostly a hobbyist/enthusiast thing.

Not true at all.. but most large object storage redundancy has moved to the application layer. There are still workloads that are a great fit but I agree OpenZFS is currently not well suited as a general purpose FS.

Not a btrfs user but OpenZFS u til recently didn’t even support O_DIRECT and may still duplicate page cache in arc. It’s not well suited as a general purpose FS where you can’t anticipate the workload, like on a desktop.

Agreed. Buy more RAM but these days that means taking a HELOC

systemd is malware.

Truth be told—- I don’t use swap and set `vm.swappiness=0` but in these difficult times, swap may be the hero we need.

That’s a good call out. I’ve never used ZFS on root. Been a ZFS user on Linux since the FUSE days but it has never seemed particular well suited (O_DIRECT support came late, ARC/page cache redundancy, etc). I can’t guarantee a consistent enough workload to optimize ZFS for and I don’t feel like it’s general purpose enough. Large object storage though? I’m sitting on a PB of ZFS.

How often are you needing to change partition geometry after install, on a server?

If you want to share step-by-step, I’ll update the gist.

Triggered

If you like that abstract geometric headline image, you might also like the source artist their model “learned” it from: https://leegriggs.com/xgen-rendered-with-arnold-for-maya

Agreed. Love stuff like this. Keep it comin’

Context: spent my career building CDNs at Cloudflare/facebook/pinterest.

The h3 juice is worth the squeeze. 0rtt TLS, BBR, HPACK, etc. You can roughly achieve this with a carefully configured h2 deployment across multiple layers and packages or just move to h3 and let the app layer handle it all. We baked all of the best practices into h3, the world just needs to move on.

Seems like the major gripe is over my usage of the word “obsolete” instead of “deprecated”. We could all be flipping switches instead of tapping on MacBooks but you gotta draw a line somewhere.

Agreed! I love naming things :D